Conventional statistics and hierarchical Bayes: a discussion on the difference in philosophy

I will join and give a talk about my recent work on conservation genetics of spruce at Institute of Statistical Mathematics (ISM) on this Thursday. I have been preparing for the seminar for this weekend, and it will take for another couple of days.

To be honest, giving a talk at ISM is the kind of work that I'd like to avoid. Unlike to a meeting of ecologists, audiences come from various scientific backgrounds, although the title of the seminar is like "statistics and field ecology". How could I appeal to ecologists and mathematicians simultaneously?

For this time, I have one idea to establish my talk. It is about differences in philosophy between conventional statistical tests and statistical models, especially hierarchical Bayes (HB) models.

Thanks to good textbooks for ecologists and softwares such as WinBUGS, HB have widely spread among ecologists for last couple of years. On the other hand, as long as I know, the most of researchers who intended to use HB did not notice to differences in philosophy between conventional statistics and HB. They thought HB was just an improvement of conventional statistics. Mathematically, it may be true for some cases. But I'd like to emphasize the difference in the way to analyze the data.

Conventional statistics such as t-test summarlize the data to simple form, like "there was significant difference between two values". Then, researchers discuss about processes of interest from those significant differences among values. In this case, data can be analyzed without any prior hypothesis.

On the contrary, statistical models such as HB fits the prior hypothesis on the process of interest (or the model) to data. Then, researchers examine the shape of the fitted model. In this case, the model should be exist prior to analysis.

If I contrast those methods in their philosophy, I would say, conventional statistics is inductive, but HB is deductive.

This difference is not serious in the simple statistical models such as linear regression, ANOVA and GLM. Since the shape of the model is limited, researchers usually do not care about the process of interest, but only care about significances of coefficients in the model. Simple statistical models can be used like conventional statistical tests.

However, as the model become flexible, researchers must pay attention to the shape of the process of interest prior to analysis. HB can model complex processes very flexibly, and can integrate multiple processes which interact each others. In this case, the significance of individual parameter is no longer important, rather the behavior of entire process is important.

Unfortunately, some researchers did not noticed this difference, and still care about the "significance" of individual parameters in the large complex model without considering the entire behavior of the model. The reason they still care about the "significance" is probably because "significance" is easy to understand, and they do not have idea to discuss the result from HB.

In my talk, I'd like to show one example to discuss the result from HB. That is quantitative prediction of the process under multiple scenarios. Actually, the result is still insufficient, and it needs more suggestions. I would be happy if I discuss about above my idea and my results there.

Conservation of Wildlife and Oil Palm (Kinabatangan River, Sabah)

If I were asked "What was the highlight of this trip to Malaysia for you?", I would answer "River cruise along the Kinabatangan". Probably, most of other participant would answer so.

During the trip in Sabah, we stayed several days at the accommodations which located beside the Kinabatangan River. And we enjoyed wildlife observation from the boat in mornings and evenings.

Thanks to skilled guides and our luck, we had chances to meet many species of mammals, birds and reptiles. Proboscis monkey, crab-eating macaque, hornbills, eagles and kingfishers were frequently observed species. Also we occasionally met orangutans, leaf monkeys, gibbons, Borneo elephants, crocodiles, etc... Accompanying people who had been working there said that we were unbelievably lucky guys to meet such species within a couple of days.

The only regret was I could not take a nice photo by myself. One of our colleagues took beautiful animal photos during the cruise. I expect he would soon upload some of them to his blog.

Anyway, it was really precious time for us.

During the stay at Kinabatangan, we learned about the wildlife conservation acts in the lower Kinabatangan which carried out by Borneo Conservation Trust and SARAYA Co. LTD.

Not only in the lower Kinabatangan, the biggest issue in wildlife conservation in Borneo island is oil palm plantation. Oil palm plantation has been spread around Borneo for last couple of decades, and it resulted in rapid destruction of rain forest and wildlife there. During the river cruise we frequently met the scenery that the oil palm plantation came close to the river which cause a significant effect for wildlife habitat.

First lecture was given by Mr. Hirotsugu Daishima from SARAYA. SARAYA is Japanese private company, and manufactures soaps from palm oil. As a pioneer of cooperate social responsibility (CSR) and wildlife conservation in Japan, they have been working for rehabilitation of the lower Kinabatangan area where oil palm plantation had once altered jungles, with Borneo Conservation Trust. Also as a member of Round table on Sustainable Palm Oil (RSPO), they contribute to the building of certification for oil palm development which minimize the damage to biodiversity and consider the welfare to the people who works in the plantation. In addition, they provide the website to appeal the such situation of Borneo's jungle and their activities for conservation, and plan study tours to Borneo for Japanese consumers.

The most impressive point in his lecture was that they were working with two opposite stakeholders: the conservationist and the industry group. Since SARAYA is a part of palm oil industry, and they need to continue their own businesses, they have involved to RSPO to realize the true sustainable oil palm plantations in tropical regions. However, they think that working only with RSPO is not preferable because major companies in RSPO still does not accept several regulations in oil palm plantation which seems to be necessary to conserve habitats for wildlife. For this reason, SARAYA made a decision to act for conservation of wildlife at the lower Kinabatangan with Borneo Conservation Trust, as long as they could.

Mr. Daishima said, "Still, we have not decided which side to take. The effort of RSPO seems to be insufficient, and it is difficult to change the mind of majority of them. But we would continue to participate with RSPO, and would like to propose for revisions of RSPO frameworks. This is our responsibility as a member of palm oil industry. On the other hand, we would continue to act for rehabilitation of wildlife habitat at the lower Kinabatangan because it is our responsibility as an palm oil company."

Next lecture was given by Dr. Toshinori Tsubouchi from Borneo Conservation Trust and Sabah Wildlife Department. As an active conservationist (also as a veterinary), he has been working in dozens of countries until now, and now is the one of the key person in conservation of the lower Kinabatangan. In the lecture, he introduced about the rehabilitation project of wildlife habitat (especially for orangutans and elephants) at the lower Kinabatangan which was carried out by Borneo Conservation Trust.

Also he gave a very impressive talk about difficulties and conflicts on wildlife conservation in developing counties. In the developed countries such as EU, USA or Japan, it is relatively easy to build a systematic strategy and consensus with stakeholders for wildlife conservation. However, because of poverty, lack of education, corruption and pressures from global economy, conservation acts are not easy as in the developed countries.

He emphasized that scientific theories does not help in the developed country for conservation, but solution is always ad hoc. It is probably true, however, non of us agreed with it. During the session, we had a very active discussion with him, but never leached to a conclusion, unfortunately.

I believe that scientists must always have roles in every conservation projects: giving a direction to the plan, developing techniques to carry out and validating the project for further improvement and generalization. But scientists usually could neither do reconciling conflicts among stakeholders nor making a decision. In other words, scientists cannot work by themselves, but always need collaboration with other activist. Probably, this is a reason why he said that scientific knowledge does not help.

We need to talk with conservation activists, and conservation activists need talk with us. Then, we need to work together. During this session, however, I could not find the way to talk and work with him. Still I do not have answer.

Probably we need to change the mind, and also they need to change the mind. The discussion with him have made a very little change to my mind to open up. I hope the discussion with us also have made him a change to his mind.

Kinabatangan River

Blue Eared Kingfisher Alcedo euryzona

Oil palm came close to the river

Conservation of Biodiversity and Sustainable Forestry in Malaysia (Deramakot Forest Reserve, Sabah)

I had tripped to Borneo, Malaysia from 23 Sep to 4 Oct for study tour of GCOE. Since it was first time to visit tropical region, everything I had met was exciting and curious. I had really enjoyed precious time.

The purpose of this trip was to visit various sites and stakeholders which involved to conservation of biodiversity and/or development of tropical forest, and to realize the current status and ongoing problems on tropical rain forests.

Here I would like to report on the trip, in several posts.

Firstly, we visited to Deramakot Forest Reserve in Sabah to learn about the sustainable forestry in topical rain forest. Owing to technical support from Germany, forest management plans were seemed to be excellent, at least better than that in Japan: intensive inventory, strict guidelines for harvest and retention of trees, well-developed techniques for reduced impact logging, etc... I believe that the systems of Deramakot should be effective for sustainable timber harvest and biodiversity conservation.

However, I noticed couple of problems. One is a lack of long-term monitoring of stand recovery after harvest. Although the damage from the harvest was very small, only four or around big trees were harvested per hectare, it should result in the change of micro-environment around the logged site such as stand structure, light condition of ground and surface soil condition. Especially, regeneration should be sensitive to such conditions. Long-term monitoring of growth of saplings and species composition should be conducted, as well as improvement of techniques of harvest and silviculture.

Another is total balance of the management unit. The reduced impact logging system in Deramakot probably requires much cost than conventional management practice. In the Deramakot system, the target area was almost perfectly monitored: species name, size, status, and even the location was recorded for all the target trees. The area was also monitored during and after the harvest to ensure that management plan was appropriately carried out. This might be additional cost to conventional management. In addition, forest certification (FSC in this case) might be another cost.

Those intensive management is possible probably because Malaysia is still developing country. Thanks to cheap salary and enough human resources, they could carry out such well-developed system. But Malaysia's economy should be grown as much as developed country near future, probably in the next couple of decades. It means that Deramakot system will financially fail no matter how it is preferable to sustainable management and biodiversity conservation.

When I asked above question to the assistant director of the site, he answered that they do not care about the cost because they are government. This answer came out probably because Deramakot system had an aspect of experiment. But developing "financially sustainable system" as well as biodiversity conservation and sustainable resource utilization is necessary to spread this system especially to the site which managed by private company. I hope they continue to improve the system to realize the "real" sustainable management system in tropical rain forest.

Plantation after harvest to assist regeneration

Harvest plan. All candidate tree for harvest is mapped.

Serial number for timber. Number is also tagged to stump for inspection.