 The title of session is Towards Criteria and Indicators for Fire Risk and Control. So we'll have two speakers in this session and we'll be talking about different aspects of concepts that relates to fire risk for peatlands. How we can map fire, how we can do the fire risk assessment, those will be covered by two speakers. Our first speaker in this session is Dr. Mothafik. He is currently a lecturer at Department of Geophysics and Meteorology at Bogor Agricultural University, IPV, in Bogor, Indonesia. He finished his PhD from Wageningen University in Netherlands and his research focuses on hydrological aspects of peatlands. His talk title today is peatland hydrological drought and fire risk assessment in changing climate. So very interesting, very topical, I would say. So Mothafik, if you're ready, over to you. Okay, Rupes, thank you very much for the time. Do you see my presentation? Yes. Can you make it a slideshow? Yeah. Yeah. Do you still have time for this webinar? Because it's already past 30, past three, right? Yeah, we are moving along, I think, we have 10 minutes behind us. Okay, everyone, good day. I'm happy to be here in the webinar. I will present my work in the past five years, it's about hydrological drought and associated peatland fires in tropical Indonesia, actually. So to begin with, I will introduce what we already have known about the degraded peatland and contribution to GHG emission in Indonesia. It's very huge carbon emission every year. And also we have a problem with biodiversity decline due to degraded peatland Indonesia. And most of the degraded peatland in Indonesia is associated with forest fires or land and forest fires. What happened if we already have what in the red picture, it's about BRG doing peatland restoration through canal blocking. We will have a different behavior of hydrological drought or forest fire in this kind of ecosystem. Therefore, we like to explore the effect of this contrasting management, either degraded peatland and derivating peatland related to fire hazard in Indonesia. My approach is we use hydrological modeling in Indonesia. It based on our last paper published in ERL, it's about hydrological drought in Indonesia. Most of water in peatland launched to drainage. So therefore, canal water level is very important aspect of hydrology in the peatland. It's controlled groundwater level in the land. Our approach is we simulate groundwater level in the land to do the changing of water level in the canal. So we compare to degraded peatland and also the derivating peatland. I tried to simulate the groundwater level and forest fire for 1980 until 2015. It's around 36 years, climate years. So I hope it also can see the climate pattern for the region. Then we focus on central Kalimantan around Insekbango to do this simulation. For fire drought index, we use a common drought index called Kichipirum drought index. We call it Kibidi. Kibidi is just a soil water depletion in peatland. It depends on the deke. Deke is a drought factor, depends on the evapotranspiration and also depends on rainfall. In my approach, because groundwater table is close to the surface, we propose to have groundwater table as a factor to reduce drought every day. Here is the formula for this approach. Then the result of my research, first we synthesis that groundwater table in degraded peatland and in rewetting peatland is quite different. You can see here in the orange like red, in red color is the degraded peatland and in the blue color is the restored peatland. This is the box plot for 1980 until 2015 period. So we make a summary of this water table. The frequency of daily groundwater table for the restored peatland is quite high. It's less than 40 cm for all of the time. But it's different with the degraded peatland. For the degraded peatland, most of the time more than 60% is below the critical threshold of 40 cm. What happened next with the fire hazard? Based on this groundwater table, then we simulate the fire hazard and compare between degraded and restored one. For the degraded peatland, the fire hazard that on the high criteria, high criteria means it's a high fire hazard. It's more than 10% whereas for the restored peatland, it's less than 80%. It's a big difference between the two. If we look at the low hazard, for the degraded peatland, it's around 91% but for the restored peatland, it's 97%. So there are 7% difference between degraded and restored. So we can use this low hazard or the hazard class to be one of the criteria to assess the biophysical attribute and peatland fire of restored peatland. From this simulation, we can see that peatland can reduce the fire risk by 30% if we compare the high hazard with the low hazard. A hazard in restored peatland compared to degraded peatland. Then we move to my other study. It's about a critical grid of the table for peatfire. We analyzed around 160 stations from BRG stations, but it's only for two years, 2018 and 2019. Baru means the new station built by BRG and the Sasami is from the Japanese government. If we use BRG station, Baru and Sasami, we can conclude that most of the time, high and extreme current above fire hazards occur when the water table is less than 40 cm. So if the water table is deeper than 40 cm, the probability of forest hazard is being high or extreme class. So therefore, 40 cm maybe is a good one indicator as a attribute for end peatland fires in the tropical peatland. So for the conclusion of presentation, monitoring ground level is crucial and technically feasible. As a part of management tools, Sasaki may be used as a reliable and fair compliance mechanism and relating peatland improve the ecosystem by minimizing fire hazards. So that's all my presentation. Thanks for the time. Thank you, Motafik. You also finished within your 10 minutes. So that's very good. So let's stay on time and I'll invite our next speaker, Dr. Solahin Manuri for his talk. And then both are focused on same topics. We can have a combined discussion and in the meantime, if there are any questions, we can address them. Dr. Manuri is a senior advisor at Diameter Consulting based in Bogor. He has more than 15 years of experience in forestry sector, including sustainable forest management, forest fire management, land use planning and forest biomass assessment, specifically focusing on peatland mapping using LiDAR and greenhouse gas emission accounting. Solahin completed his PhD from Fenner School of Environment and Society at the Australian National University. And today he will be talking about mapping fire. Can spatially explicit criteria and indicator be developed in this context? So over to you, Solahin. Thank you, Rupesh. Good morning, good afternoon, good evening, everyone. Thanks for introduction, Rupesh. So today I'm going to talk about the fire mapping and thanks for the invitation. So my my title is given by organizers, so I will try to follow the expectations. So I hope it can meet your expectations. So let me share my screen. Can you see it now? Yes, you put it on screen show mode. Yes, it's working great. Yeah, so this is the title that the organizer asked me to contribute. So yeah, maybe I'll try to do my best to meet this topic. And yeah, you know that in the past four decades, what fires have become become a common event and repetitive in the tropical ecosystem of Indonesia. And this is due to because of, you know, this prolonged drought season in Indonesia normally caused by the El Nino, certain and oscillation or and so, and also the Indian Ocean default mode. Yeah, also a result of the unsustainable practice of forest and land management as previous presenters mentioned earlier. And during the dry season become more susceptible to fires and normally it contributes to which amount of greenhouse gas emissions. And the impact of the fires in Indonesia is unfortunately is quite limited, limitedly explored or with relatively high uncertainty in particular regarding the size of the burn areas. So in this presentation, I would like to present some initiatives related with fire mapping in Indonesia and also in the tropics. As you know that the biggest challenges in mapping fires in Indonesia or in tropical region is because of cloud resistance, which is common in tropical region and also smoke and haze occurrence during the fires. And also the short windows of opportunity after the fire season and which is started during the rainy season, but normally during the rainy season is also cloudy and once we have clear sky, the vegetation is already growing. So this is what I mean about short windows of opportunity. But it's really important to have a really reliable remote sensing data. So as you can see in this, this is from Modi satellite in over central Kalimantan during 2015 fires, and you can see most are the wind coming from the southeast direction. And you can see that most of the fires are covered by smoke and haze. So this is the challenges, biggest challenges we have so far. And here is the chart about the existing initiative on fire mapping in Indonesia and also in global tropical, pan-tropical. It started in 1992, 1993 fires, Lennards and Panzer did some quite huge studies on fire extension or fire mapping in Kalimantan using Landsat imageries. And then in 1997 again big fires, bigger than the fires in 1982-1983, Sigurd and Hoffman also doing the fire mapping using a radar which is, you know, the passive sensor that can penetrate clouds. And in the meantime, there is also another database on burn area provided by Sigurd, now it's version 4.0, they already start the mappings in 1995 actually, but they already include the most recent Modi's burn scar area until recently. And then in 2000s, the Modi's burn scar are already, so the Modi's satellite is already operational and there are two types of burn area datasets that are provided by Modi's NASA. They call it MCD-45 and MCD-64. So MCD-64 is the latest version, so it is commonly used by any studies in global and also at national level. So this defect 4.0 also incorporated the burn area MCD-64A1. And then in 2006 the Ministry of Environment, sorry, actually Ministry of Environment started the mapping in 2016 to map the fires even in 2015. So where the huge El Nino was occurring. And now they are planning to extend the period until 2006 to, this is the, in order to estimate the emission from pit fires for developing the reference level for red plus activities. So they are doing, it's not published yet, but the one they published is in 2016 or fire in 2015. And at the same time, there is another studies in Indonesia about fire mapping using radar satellites. So this is the Sentinel-1, so this is a passive active sensor. And also LAPAN do the three interiors in 2015, 2016, and another one in 2019 also doing the mapping using combinations of Landsat and also Sentinel-1 hybrid between optical and active satellite. And yeah, also Copernicus started in 2014 until recently. So yeah, you can see there are four that is continuously, continuous initiative in fire mapping in Indonesia. This is the detail of the methods that the summary of the method that they use. So mostly using Landsat imageries and also radar satellite. Yeah, and also MODIS, they have a lot of global scope for global coverage. They mostly using MODIS satellites. Yeah, and the most recent one is Sentinel-1. So they already using Sentinel, which has two different sensors, optical and also active sensors with wide, often repetition from about five to 12 days. So this is quite promising and most of them are using either object-based image analysis or pixel-based classifications and also visual classification. So you can see the Ministry of Environment and Forestry, they are using the physical visual classification of Landsat imageries with the aid of other auxiliary data. Like hotspots and also verified by ground data and high resolution imageries. And this is the one that data that we can compare in fires in 2015. So you can see from GVET, MODIS, Bernskar, and also studies from Low Burger in 2017, Ministry of Environment and Forestry, and also LAPAN. So LAPAN is a space agency in Indonesia. And the GVET, MODIS, and most of them has data on feed-length fires. So, Lihun, you wrap up really quickly, your time is almost over. So this is basically the comparisons of the, and yeah, this one is the summary, maybe I can skip that. And also we have a pro and cons on the sensor type resolution and also classifications. As you can see in the previous slide that both sensors are used either passive or active and LAPAN even using the hybrid using both sensors. And for resolution, most of the national coverage, they use medium, at least medium resolution from 10 meters to 30 meters. So I think that is better than the low resolutions. And for classifications, yeah, they use manual and also automatic classifications. Yeah, this is the type of satellites we can use. And this is also the lifespan. This is very important when we select the satellite imagery we want to use. So not only because of free, but also the lifespan. So some of them already not operating, so we cannot have the historical or future data for fire mapping. So this is very important to have better knowledge about the lifespan of the satellite. And one more important thing is the accuracy assessment. So this is very important to do. Some are using like high resolution imageries, like the modis, the 500 meter resolution modis using 30 meter lens set for four validations. 25 meter maps validated using the ground measurement and ORS RL survey. So this is something that the mapping should be part of the mapping initiatives. And apart from that, also burn pit depth mapping is also one thing that very important, especially when we want to estimate the emissions from pit fires. So accurate information is really important. You have to wrap up in 30 seconds at the most, otherwise we won't have time for the rest of the day. This is the last slide. So main message is basically we have already initiatives on fire mapping. And especially explicit criteria should be developed depending on the methods and data they use, combined with knowledge from a ground and IRL survey. So I think it is one thing that really important validations and also selections of methods should be depending on the mapping coverage, reporting interval and technical capacity. So a lot of variations can be used depending on the needs. Yeah, I think that's all my presentation. Sorry for taking a longer time. Thank you, Rupesh. Thank you, Salahin. Thank you for sharing. Your presentation was quite informative with a lot of very, very interesting and useful information. I'm sorry we have only limited time, but you did provide a good flavor of where this mapping fire science stands. So just to very quickly wrap up this session on fire was very useful in the sense that we kind of get the two extremes. Well, first speaker Muthafik talked about this monitoring of groundwater table and then, you know, like right at the level of location, like, you know, in the physical in the field field setting, what is happening there and getting the information from that level. While Salahin talked about more from a very kind of a top down if I can use the term with the radar and those kind of technology, how you can do the large scale regional mapping. And both of these are quite important, like in terms of validating the data which is collected at a very broad scale. It becomes very important, like having boots on the ground, someone measuring someone collecting the data in the field. And I think that the interesting and important aspect is how do you sort of stitch them together into useful relevant criteria and indicators. So they are useful for people who are practitioners who are has the responsibility of making some decision based on that. So this ties together very nicely with the previous two sessions where it has to be thought up from what are your principles and then criteria and indicators and all speakers have presented very detailed approaches when it comes to biophysical aspects. There are so many nuance that we need to know their scientific approaches, there are technology, there are advancement, but then it becomes what is cost effective, what is good enough. So I haven't seen any specific question coming for this session. There was one in Basha Indonesia which I requested Daniel to translate and that was for Tophik. That's right, this is for Tophik. Tophik's question is about ground water level you presented. How can you use that information to estimate greenhouse gas fluxes? The second question is still related to your categories of pitland. This is from Pablo Martin. How do you categorize pitland is restored and how is it degraded? What kind of criteria you use? Okay, thanks Daniel. For the first question we can use ground water table as a proxy for greenhouse gas emission as already mentioned, Professor Makrit I think from your question. We can use ground water table as a proxy for instance through, sorry, for your approach for instance to estimate greenhouse gases. For the second question from Pablo, thanks. How to make a classification of degraded and restored? Our approach is just simply if we have a canal brooking or just a restoration project on there, we will simply, it is a riveting project, so it's riveting pitland. Otherwise, if it is a degraded, that's simplification. Thanks, Pablo. Thank you, Tophik. Are there any other questions? Daniel, did you notice something else that we need to address? Yes, it's very general questions. Maybe anybody can answer that even from the previous session. It's about the term use as sponge cities. Is it something to do with pitland or is there any merits about categorizing city as sponge city? Yeah, if you want to respond, just unmute yourself and you can respond. Well, while the speakers are thinking, it's another one I picked about the canal blocking. Again, I think any of those who speak about canal blocking can respond to that. There is a negative impact also on blocking the canal because it will raise, sorry, it will lower the pH to some extent. Any comment on that? Yeah, this relates to, I think, the phygeochemistry of soils, how the redox changes and what impacts it has. So any panelists want to take this? Rubik is still. Sorry, I haven't been working in that area. I don't see really how immediately canal blocking can have an effect on the pH. Maybe this takes some several years. So it's a bit of a tricky question, I would say. Yeah, and it could also be that there would be some pockets where you could have this effect and then there would be, you know, at a landscape level there would be some areas which are not that affected. There was one more, I think, question or comment about the behavior change in terms of fire because sometimes the soil is dry and it is very vulnerable, but sometimes it's a human action that caused the trigonite. So that is a point very well taken, but there are also natural fires with lightning and all that. So it has to be taken into account that if a peatland is vulnerable because of less water, then it's not only always just human action that the risks are there, of course. Okay, so I think there is no comment about cities unless you want to say anything partennial. Otherwise, I think we in the interest of time, I know you have to wrap up and we have to have closing remarks from you. So I will yield the time to you. Okay, thank you. Thank you very much, Rupesh, for sharing very interesting and lively session. We have 172 people in the room, virtual room, and they are very active and watching us. Suddenly, the five speakers have been bringing us an incredible travel right from the sky in the outer space from the satellite down to earth, not only on the ground, but also going down below the ground to look at groundwater table, biochemistry, etc. So it's been very rich conversation we have, and this challenge us to really think about how to use this kind of information, which we call technology, Ravi characterizes criteria into the people we are working for. And it's been very clear about that how to make these people centered. So this information can be gathered getting more and more easily today's and how to bring that to communicate that with people in the context of pit restoration. So the way I see it is really challenging to look at our information that we can generate very quickly and perhaps more and more accurately means something for people. So bringing this into people, you know, conversation day to day conversation is not easy. This is going to be a challenge how when we are going to talk with them and verify this kind of thing. And what does it really mean for them. So today we learn a lot from this conversation about those criteria that's been described from the hydrology from the emission of greenhouse gases fire hazard and risk. Again, what does what do they really mean for people. So we have a big challenge in front of us and certainly this will be a very useful way of guiding us towards, you know, verifying these numbers and criteria that is generated. So the indicators that Ravi underline as as information will be will be based on data. So data is very useful. It's, it can be easily generated now even if you do it through a virtual way by synthesizing or assessing reviewing papers that can be generated but again, what does it mean for people. So with that, I guess we have a lot of challenge in front of us to do. And at the same time was also challenged by Harris about people around us about their agenda. So that might be our next round of webinar or discussion to have. So thank you very much again for those who've been very patient and sincere and and also based on to follow this conversation and also contributing person and chat. Thank you for your kind contribution. And with that I would like to close this webinar today. And again for the speakers we will be having our debriefing session in another room. Thank you.