Ecological restoration of the Philippine dipterocarp forest ecosystems : the role of spatial, meso-scale climatic modelling

Abstract

The rich and diverse dipterocarp forest ecosystems of the Philippines provided significant volume of commercial timber and have made a substantial contribution to the country's economy for many years. Dipterocarp forest ecosystems dominated the forest landscapes, however, over the years intense land-use pressures have led to their degradation and fragmentation. The future sustainability of the tropical forest ecosystems is recognised to depend largely on the ecological restoration of degraded and human dominated forest landscapes. The overall goal of this research was to investigate the issues and challenges facing the restoration of the Philippine dipterocarp forest ecosystems and in particular, examine the role of spatial, meso-scale climatic modelling. Ecological restoration is a process of bringing back a degraded system to a condition that is desirable, healthy and fits into its natural surroundings. Driven by the increasing recognition of ecosystem management, ecological restoration has emerged as an important scientific approach in landscape conservation. However, as the focus of ecological restoration gradually shifts from a species/site specific approach to an ecosystem-based approach, two key challenges in restoring forest ecosystems are emerging. First, the protection of remnant forest ecosystems in view of their complexity and the risks and uncertainty associated with their restoration. Second, the need to develop land-use strategies that will (1) lead to the expansion of remnant forest ecosystems, (2) reduce pressures on whatever is left, and (3) complement human needs for ecosystem goods and services. One of the ways to meet these challenges is to gain a better understanding of the complex suite of environmental processes that underpin the maintenance and restoration of forest ecosystems. Such an understanding is necessary for designing strategies that will facilitate, accelerate and maintain the development of the various stages of forest succession and inform restoration planning at the landscape level. In choosing an appropriate scale of analysis, regard was given to the environmental processes controlling the distribution of resources critical to the life-support system of forest ecosystems and the hierarchy of scales in which they occur. One ot the scales by which these processes can be observed or examined is the meso-scale. Meso-scale climate applies to marine, continental, desert or mountain environments and to a small country or region. At the meso- scale, the influence of climate is one of the dominant environmental processes. Climate determines the amount of precipitation and insolation as well as the levels of air temperatures that are vital to the growth and development of ecosystems. There are several advantages in focusing on the meso-scale. The constraining influence of meso-scale processes on the growth and development of forest ecosystems is well recognised, but has been poorly integrated in planning for restoration and rehabilitation - in no small part due to the unavailability of appropriate tools to assist the decisionmaking process. Developments in environmental modelling now provide the capability to effectively deal with meso-scale processes, particularly those influenced by topography on climate. There is also a match in environmental process and scale of investigation. Restoration of forest ecosystems is a landscape conservation issue and meso-scale processes can be observed at landscape level involving thousands of hectares or over a small country or region. The meso-scale climate modelling approach can contribute to understanding the complex nature of ecosystem-environment relationships. With the aid of modelling tools and computer-based technologies, understanding of environmental processes that are critical to the life-support systems of forest ecosystems are taken beyond the traditional and common scales of investigation. The long-term monthly mean meso-scale climatic models for the Philippines were developed using the spatial interpolation package ANUSPLIN. The ANUSPLIN package is based on the technique of thin plate smoothing splines and is used to fit surfaces-to noisy climate data as a function of one or more independent variables. Crucial to the accuracy of the models is the incorporation of elevation in the spatial interpolation of monthly mean data. The errors of the models are within the acceptable limits when interpolating climate data from a sparse meteorological network. The predicted standard error for both the maximum and minimum temperature models were about 0.5°c. For the precipitation models, the standard errors ranged from 10-20% of the network mean. The spatial distribution of the standard error highlighted parts of the landscape where additional stations could improve the accuracy of the models. For temperature models, these areas included the western highlands, eastern ranges and northern lowlands of Luzon, western Panay and Negros in the Visayas, and large parts of Mindanao. For the precipitation models, interpolation could be improved if additional stations were available along the eastern side of mainland Luzon, Mindoro, Palawan, Panay, Catanduanes, Samar in the Visayas, and the northeastern, central, and southern parts of Mindanao. The models revealed important changes in the Philippine climate in the 20th century. The estimated monthly mean surface values suggested a general pattern of declining precipitation. The change in precipitation was significant in the dry season from January to May and in July and September. Examination of the difference in precipitation distribution for January and July showed where the changes occurred and by how much. The January model showed that the average rainfall for this month decreased by up to I00 mm across the plains and valleys. However, other isolated places gained precipitation including the northeastern coasts of Mindanao (500 mm) and up to 200 mm in coastal and mountain ranges. In July, the change appeared along the western side where large areas experienced decline reaching up to 400 mm. The mountain ranges and coastal ranges gained up to 200 mm. It is important to note that some of the extreme changes may be due to the poor precipitation data network. However, the high incidence of drought and ENSO events affecting large parts of Asia support what the models suggest as the growing dry conditions across large parts of the Philippines during the second half of the century. Overall, the spatial models appeared to have captured the variability in the Philippine climate that is due to the influence of the changing wind systems, tropical cyclones, trades and localised disturbances and their interactions with the local topography. Meso-scale climate analysis aids in understanding the influence of climatic processes that are critical to the basic functioning of forest ecosystems. Using the ANUCLIM package and the BIOCLIM procedure, the long-term meso-scale spatial models were integrated with the distribution data for seven dipterocarp species to determine their potential climatic domain. The range distribution of each species was based on the minimum and maximum values of the bioclimatic parameters while the core distribution was determined by values of bioclimatic parameters within the 5-95% percentile limits (i.e. the climatic conditions which encompassed most of a species distribution data). These range and core distributions were mapped on a regular grid. The models showed that the low-lying areas that experienced decline in precipitation in the latter half of the 20th century do not represent the predicted range and core bioclimatic domain of dipterocarp species including the majority of the restricted endemic and non-endemic species with limited distribution. Although the predicted range covered broad geographical space for some species, the predicted core areas emerged as highly dispersed and spatially restricted. The predicted core areas covered the narrow stretch of landmass connecting the southeastern part of mainland Luzon to the Bicol Region and mountainous provinces. They can be considered as the remaining potential lifelines of the dipterocarp forest ecosystem to build upon. The core areas, however, were fairly narrow, isolated, and mostly running longitudinally demonstrating their vulnerability to various forms of disturbance. The core areas also captured the meso-climatic habitats of dipterocarps considered to be restricted endemics and nonendemics with limited distribution. Analyses of the models in conjunction with current land-use revealed considerations important for planning restoration projects. About 21 % of the remaining forest cover coincided with the predicted core areas. This finding implies that these remnant forest are likely to contain dipterocarps and support these species in the future based on the influence ·of meso-scale climatic processes. The predicted core areas that are forested run along the eastern coasts of Regions II and IV. The mature dipterocarp forest in these regions have an important role in ecological restoration. There is growing evidence that the timing of mass flowering common among dipterocarps is linked with El Nino or La Nina. El Nino event are believed to be occurring more frequently in the last three decades. If it is true that this event triggers the flowering and fruiting of dipterocarps, then the mature forest that contain dipterocarp species could become the most important sources of regeneration materials for restoration. The report by the Philippine government revealed the poor supply of dipterocarp seeds as a consequence of which limited areas have been planted with these species since the start of its reforestation programme about 20 years ago. These findings imply that the allocation of dipterocarp seeds would require careful planning and the sources of these seeds must be secured to ensure that future supply is not further compromised. The regions with mature dipterocarp forest, therefore need to be managed as priority areas for protection because they contain the potential forest ecosystems around which large-scale restoration projects can be based and are the future sources of seed for regeneration of dipterocarps elsewhere. The findings also showed the lack of reforestation programs in regions where the dipterocarp species are likely to be found and where existing forests are already in place. The opportunity to reduce pressures on remnant forests particularly those surrounded by relatively new or limited disturbed areas, may have been missed by moving reforestation programs away from these places. About 20% of the predicted core areas lie along the fringes of remnant forests classified as brushland. Brushland is not entirely devoid of forest cover and the forms of vegetation in these areas represent stage/s in forest succession. The proximity of brushland in forested areas is an opportunity for ecological restoration because these areas can be utilised to link remnant forests to those areas that could be established. Brushland can be prioritised as target reforestation sites depending on the nature of degradation and vegetation, adjacent land-use, and their potential to support the growth and development of dipterocarp species. The overall representation of protected watershed forest reserves in Regions II and IV is poor. The national parks in these regions, however, are larger than in other regions and they are also known to support dipterocarp forest ecosystem. The national parks in Region V in Luzon and parts of Mindanao are known to support dipterocarp forest ecosystems. The predicted core areas captured some of these parks. These findings are useful considerations in the improvement of the protected area system. Other core areas have limited forest cover and are dominated by agricultural activities, hence their potential in ecological restoration is limited. A critique of the Philippine Master Plan for Forestry Development revealed the lack of a systematic ecological restoration plan for the dipterocarps, the inadequacy of the protected area system, weaknesses in the current forest management strategies, and most importantly the increasing uncertainty in the future supply of regeneration materials for ecological restoration. This research has shown a new perspective of opportunities that can contribute to the development of an integrated national ecological restoration plan. These opportunities for restoration lie largely along the areas adjacent to the eastern seaboard. These areas contain the potential remaining lifelines for the dipterocarp forest ecosystems and the best opportunity for the restoration of dipterocarp forest ecosystems through various mechanisms of land-use interventions. However, there are no easy opportunities and hard decisions have to be made to take national restoration efforts forward in what will be a difficult, long, complex and uncertain journey. Show more