El Nino x forest resilience

Lead Research Organisation: Imperial College London
Department Name: Life Sciences

Abstract

The widespread and rapid logging of tropical forests may mean they are unable to withstand climate shocks such as El Niño events.

Logging slashes the abundance of invertebrates like ants and earthworms, but vertebrates can take up their roles in the ecosystem. Invertebrates perform essential functions for the smooth running of the ecosystems in tropical forests. For example, creatures such as termites and millipedes help dead leaves decompose and release their nutrients back into the soil, and carnivorous ants and spiders act as predators of herbivorous invertebrates that would otherwise munch through all the foliage.

Nearly a half of all tropical rainforests worldwide have been logged, and this often causes heavy changes to the number and type of invertebrates, with many species being lost from the ecosystem. But when invertebrate diversity declines, their vital functions can be carried out by other members of the ecosystem, including vertebrates including rodents, birds and bats. Because of this, rainforest ecosystems have a remarkable resilience to change.

However, although the ecosystem can continue to function with vertebrates taking more of a leading role, the situation leaves rainforests vulnerable: the forest will keep maintaining itself, but it will be much more susceptible to further change. Relying on vertebrates is probably a bad tactic - they are less diverse and vulnerable to new challenges.

One such new challenge is the drought that accompanies El Niño events. There is already evidence that El Niño events can disrupt the resilience of ecosystem processes to logging disturbance. For example, the number of seedlings germinate and establish in logged forest during an El Niño year is just one-sixth of the number that establish in primary forest during the same year. What's more, those seedlings then die almost twice as fast in a logged forest during El Niño than they do in primary forest.

This project will examine the resilience of tropical forest ecosystem processes to an El Niño event, asking specifically whether El Niño reduces the resilience of ecosystem processes to logging disturbance that otherwise keeps those forests functioning between El Niño events. I will do this by comparing historical data collected in both primary and logged forests with new data that will be collected during the 2016 El Niño event. My goal is to generate a broad picture of the interactive impacts of El Niño and logging on the forest ecosystem as a whole, for which I will necessarily sacrifice detailed understanding of any one ecosystem process.

Instead, I will be looking at a very broad range of data types and measurements of how the forest works and what species are doing the jobs. The data will record the rate at which 15 ecosystem processes operate in primary and logged forest habitats, encompassing metrics of tree dynamics, interactions across plant and animal communities and nutrient recycling processes. These data will be compared to records on the abundance and biomass of 26 functional groups nested within seven plant and animal communities, covering all major trophic groups including early and late successional tree species, and vertebrate and invertebrate herbivores, predators, scavengers and omnivores.

Planned Impact

The beneficiaries of this research fall into four broad groupings, the first three of which are interrelated and stand to gain similar benefits.

1. Public sector organisations, particularly the Sabah Forestry Department (SFD): SFD is responsible for the sustainable management of Sabah's extensive forest estate, issuing all logging concessions and permits for conversion of logged forest estates into plantation. I work closely with SFD on the SAFE Project and SAFE has been strongly supported by Datuk Sam Mannan, the SFD Director. I will be able to feed results and management recommendations to Datuk Mannan and the relevant management committees with the potential to directly influence SFD policies and requirements for new logging and land use clearance operations.

2. Private sector, specifically oil palm and forestry companies: This study is based at the SAFE Project, a collaboration between academia (led by myself) and the oil palm industry (particularly Sime Darby, the primary funder of SAFE). This research will help quantify the resilience of logged forest patches that are retained within oil palm plantations - which have almost inevitably been logged - to extreme climatic events such as El Niño. It has considerable relevance to the design of new plantations in which retaining High Carbon Stock forests is rapidly becoming an important consideration. Data collected in this project will also have relevance to designing best-practice Reduced Impact Logging techniques and thresholds of logging intensity beyond which forests lose resilience, and will be of interest to companies such as Usahawan Borneo, the logging company contracted to work at the SAFE Project. I will be able to feed results to the Board of Sime Darby and to the CEO of Usahawan Borneo, along with any recommendations about best practice.

3. Third sector organisations, particularly conservation NGOs and industry groupings such as the Round Table for Sustainable Palm Oil (RSPO) and the High Carbon Stock (HCS) consortium: I will work closely with the South East Asia Rainforest Research Partnership (SEARRP) that supports scientific research in Sabah, including the SAFE Project, and that has two dedicated Impact Managers whose role is to translate science into policy. SEARRP holds a seat on the RSPO Biodiversity Committee and is well placed to feed results from this proposal into this high-level grouping of industry and conservation NGOs. My results will have important implications for the definition of 'sustainable' landscapes and plantations that form a central component of oil palm certification schemes operated by RSPO and the scheme currently in development by HCS. I sit on the Science Advisory Committee for HCS and will be able to feed results directly into their decision making processes.
4. Academic community: The researcher community will benefit from the new knowledge and extensive datasets generated by the project and the scientific advancement on a topic of global importance. All data generated in this research will be made available via the NERC Environmental Information Data Centre and the SAFE Project online database. My proposal will present the broadest assessment of forest ecosystem resilience to El Niño that has ever been conducted, providing a unique opportunity to examine the multitude of linkages within and among functional groups and ecosystem processes.

Publications

10 25 50
 
Description We detected an economically important species of tree may be experiencing reproduction failure.

We also found that severe drought, such as that experienced during El Nino events, challenge the process of natural forest regeneration in logged forests. In heavily logged forests, seedling mortality was skewed towards primary forest species, meaning the process of succession was reversed, whereas primary forests were more robust to the drought event.
Exploitation Route Our findings could help direct tree conservation efforts, targeting the early seedling stage of Bornean ironwood and other key primary forest species
Sectors Environment

 
Title Impact of El Nino drought on seedling dynamics 
Description In each of the 84 SAFE vegetation plot a 5 m x 5 m seedling plot was established in 2012. All tree and liana seedlings >= 50 cm height and <= 1 cm DBH were tagged mapped and identified. At the centre of the 5 m x 5 m plot a subplot of 2 m x 2 m was marked out where all seedlings >= 10 cm height were included. Censuses 1&2 in 2012 were carried out by Hamzah Bin Tangki (PhD thesis 2014); census 3 in 2015 was carried out by Michael Massam (MSc thesis 2015); census 4 in 2017 was carried out by Elizabeth Telford. Diameter was measured at 20 cm from ground. Height was measured for tree seedlings (not measured if stem was snapped). A number of stems are missing diameter and height records in earlier censuses but considered alive at those times based on information from census 4 -- these data rows are kept and NA values assigned. Species id were done by SAFE field botanists. 2017 recruits stems have less complete identification. 11 seedling plots were not measured due to disturbances, mainly logging road and landslide. At census 4, for each seedling in the plots which was deemed big enough, the fifth youngest leaf counting from the top shoot was collected. If damaged, the nearest alternative was taken Leaf petiole was removed and leaf was photographed while fresh. Fresh weight and oven dry weight were measured. Leaf area was calculated from photographs using ImageJ. These measurements were used to calculate seedling CSR trait values (Piece, S., et al. 2017 Functional Ecology) 
Type Of Material Database/Collection of data 
Year Produced 2018 
Provided To Others? Yes  
Impact NA 
URL https://www.safeproject.net/datasets/view_dataset?id=178
 
Title Increased importance of terrestrial vertebrate seed dispersal in tropical logged forests 
Description Description: A large seed dispersal experiement combining seed tracking and camera trapping at ten forest sites along a wide gradient of historical logging disturbance with AGB ranging between 4.7 and 614.0 Mg ha-1, all part of the established SAFE mammal survey network. Each experiment was run for a consecutive five days using experimental seeds with different hardness (fleshy vs hard) and size (large vs small). Each seed was tracked with a spool. Project: This dataset was collected as part of the following SAFE research project: Resilience of Tropical Forest Ecosystem Processes to the Interactive Effects of El Nino and Forest Disturbance Funding: These data were collected as part of research funded by: Natural Environment Research Council (Directed grant, NE/P00363X/1, https://gtr.ukri.org/projects?ref=NE%2FP00363X%2F1) This dataset is released under the CC-BY 4.0 licence, requiring that you cite the dataset in any outputs, but has the additional condition that you acknowledge the contribution of these funders in any outputs. XML metadata: GEMINI compliant metadata for this dataset is available here Files: This consists of 1 file: template_seed_experiment_LanQieJan9.xlsx template_seed_experiment_LanQieJan9.xlsx This file contains dataset metadata and 2 data tables: Seed fate (described in worksheet Seed fate) Description: seed fate and removal distance of 12000 experimental seeds of different treatments at all sites Number of fields: 14 Number of data rows: 12000 Fields: seed.id: Seed ID (Field type: ID) Grid: Experimental site ID, with the same Grid identifier used in the core SAFE project small mammal trapping work -- see SAFE dataset 256 "CORE SAFE PROJECT SMALL MAMMAL TRAPPING DATA" (Field type: ID) Point: experimental points, selected from the camera trap points in SAFE gazetteer (Field type: Location) Day: The day of the seed outcome. Each experimental point was set up on Day 1 and checked on Day 2-6. This numeric variable is used for temporal analysis. For treating Day as a random effect, an additional "date" label can be created by pasting Grid and Day. (Field type: Numeric) type: Experiment seed type. OP = oil palm fruit, PK1 = single pumpkin seed, PK10 = pumpkin seed cluster of 10, PK20 = pumpkin seed cluster of 20, PT = pistachio nut. For analysis, these were classified as fleshy (OP) vs hard (other seeds), and large (> 10 g; OP, PK10, PK20) vs small (< 10 g; PT, PK1). (Field type: Categorical) trmt: Experiment treatment. Control = seeds accessible by all animals, Cage = exclosure cage treatment with 10x10cm entrances where large vertebrates were excluded (Field type: Categorical) Distance: Seed removal distance (Field type: Numeric) Bearing: Compass bearing of the removed seed from experimental point (Field type: Numeric) Location: Location of removed seed. Free text can be grouped into categories for analysis. (Field type: Comments) fate: Seed fate. Untouched = intact and not moved. Uneaten = removed but uneaten (dispersed). Eaten = eaten or partially eaten. Unknown = seed dragged into burrows, nests or up trees with seed fate unknown, presumed eaten in analsysis to be conservative about seed dispersal (Field type: Categorical) Day.3: Fate of dispersed seeds on subsequent days, Day.3 - Day.6, where applicable, e.g. a seed dispersed on Day 4 would only be monitored on Day 5 and Day 6. Also, not all dispersed seeds could be practically monitored. (Field type: Categorical) Day.4: Fate of dispersed seeds on subsequent days, Day.3 - Day.6, where applicable, e.g. a seed dispersed on Day 4 would only be monitored on Day 5 and Day 6. Also, not all dispersed seeds could be practically monitored. (Field type: Categorical) Day.5: Fate of dispersed seeds on subsequent days, Day.3 - Day.6, where applicable, e.g. a seed dispersed on Day 4 would only be monitored on Day 5 and Day 6. Also, not all dispersed seeds could be practically monitored. (Field type: Categorical) Day.6: Fate of dispersed seeds on subsequent days, Day.3 - Day.6, where applicable, e.g. a seed dispersed on Day 4 would only be monitored on Day 5 and Day 6. Also, not all dispersed seeds could be practically monitored. (Field type: Categorical) Camera trap records (described in worksheet Camera trap records) Description: For each visit to seed experiment by animals recorded by camera traps, we recorded the functional group (large vertebrate or small vertebrate) and seed activity (eating or moving) Number of fields: 12 Number of data rows: 2594 Fields: Grid: Experimental site ID, with the same Grid identifier used in the core SAFE project small mammal trapping work -- see SAFE dataset 256 "CORE SAFE PROJECT SMALL MAMMAL TRAPPING DATA" (Field type: ID) Point: experimental points, selected from the camera trap points in SAFE gazetteer (Field type: Location) Day: The day of the camera trap record. Each experimental point was set up on Day 1 and checked on Day 2-6. This numeric variable is used for temporal analysis. For treating Day as a random effect, an additional "date" label can be created by pasting Grid and Day. (Field type: Numeric) trmt: Experiment treatment. Control = seeds accessible by all animals, Cage = exclosure cage treatment with 10x10cm entrances where large vertebrates were excluded (Field type: Categorical) func.group: functional group of the seed visitor (Field type: Categorical) aminal.comment: animal species if possible to identify on photos, with some level of uncertainty (Field type: Comments) visit.number: visit number of the apparent repeated visits by the same animal, with some level of uncertainty (Field type: Numeric) PIT.tag: the PIT tag number of tagged animals entering cage, detected by the antenna and recorded by the data logger (Field type: ID) estimated.body.size: body size estimate from photos, in mm, with some level of uncertainty (Field type: Comments) activity: observed interaction with seeds, with details in the next column. For analysis, "investigate" was not considered an active interaction. (Field type: Categorical) activity.comment: detailed comments on the activity (Field type: Comments) seed.type: the seed type(s) interacted with, if possible to determine, with some level of uncertainty (Field type: Categorical) Date range: 2017-03-01 to 2017-10-31 Latitudinal extent: 4.6881 to 4.7519 Longitudinal extent: 116.9633 to 117.5934 
Type Of Material Database/Collection of data 
Year Produced 2019 
Provided To Others? Yes  
 
Title Increased importance of terrestrial vertebrate seed dispersal in tropical logged forests 
Description Seed fate and removal distance of 12000 experimental seeds of different treatments at all sites. For each visit to seed experiment by animals recorded by camera traps, we recorded the functional group (large vertebrate or small vertebrate) and seed activity (eating or moving). 
Type Of Material Database/Collection of data 
Year Produced 2019 
Provided To Others? Yes  
Impact NA 
URL https://www.safeproject.net/datasets/view_dataset?id=276
 
Title Spatial aggregation and seedling survival of the Borneo Ironwood (Eusideroxylon zwageri) 
Description Description: This dataset was collected by the Imperial College London MRes Tropical Forest Ecology Field Course as a group project in Maliau Basin. Fieldwork was carried out on 29 Jan - 1 Feb 2018. A 4-ha plot was set up to assess the spatial aggregation of the Borneo Ironwood (Eusideroxylon zwageri, local name Belian) and seedling survival. All trees with DBH >1cm were mapped with XY coordinate rounded to nearest 1m. DBH class was rounded down to the nearest 10cm, e.g. size 10, 20, 30 etc. Size class 1 refers to trees with DBH between 1 and 10cm. A swamp area on the north and northeast of ~1-ha was not assessible hence data defficient, but as Belian is not known to growth in water logged conditions the species was presumed to be absent in this swamp area. Belian seedlings were surveyed in 540 5x5m subplots in the centre zone of the 4-ha plot. In each seedling plot, healthy and damaged (primarily herbivory damage by mammals, defined as damaged when top shoot, i.e. apical meristem, was missing). Other variables recorded in seedling plots include slope angle and ground vegetation cover. Project: This dataset was collected as part of the following SAFE research project: MRes Tropical Forest Ecology Field Course Funding: These data were collected as part of research funded by: Natural Environment Research Council (Directed grant, NE/P00363X/1, https://gtr.ukri.org/projects?ref=NE%2FP00363X%2F1)Imperial College London (MRes Tropical Forest Ecology field course program)This dataset is released under the CC-BY 4.0 licence, requiring that you cite the dataset in any outputs, but has the additional condition that you acknowledge the contribution of these funders in any outputs. XML metadata: GEMINI compliant metadata for this dataset is available here Files: This consists of 1 file: Belian_data_for_SAFE_upload_LQ_10Jan_2019.xlsx Belian_data_for_SAFE_upload_LQ_10Jan_2019.xlsxThis file contains dataset metadata and 2 data tables: Belian tree coordinates (described in worksheet Belian trees)Description: All trees with DBH >1cm were mapped in a 4-ha plot with XY coordinate rounded to nearest 1m. Each row corresponds to a Belian tree. DBH class was rounded down to the nearest 10cm, e.g. size 10, 20, 30 etc. Size class 1 refers to trees with DBH between 1 and 10cm. A swamp area on the north and northeast of ~1-ha was not assessible hence data defficient, but as Belian is not known to growth in water logged conditions the species was presumed to be absent in this swamp area.Number of fields: 9Number of data rows: 91Fields: X10: X coordinate of the southwest corner of 10x10 m subplot (Field type: Numeric) Y10: Y coordinate of the southwest corner of 10x10 m subplot (Field type: Numeric) X: X coordinate within the 10x10m subplot (local X) (Field type: Numeric) Y: Y coordinate within the 10x10m subplot (local Y) (Field type: Numeric) class: DBH rounded down to the nearest 10cm, e.g. size class 10, 20, 30 etc. Size class 1 refers to trees with DBH between 1 and 10cm (Field type: Numeric Trait) remark: remark on whether tree is a resprout from the base of an older tree and other tree condition information (Field type: Comments) collected_by: members of field teams: 4 groups of 4 students each (Field type: Comments) X_coord: X coordinate within the 4-ha plot (global X) (Field type: Numeric) Y_coord: Y coordinate within the 4-ha plot (global Y) (Field type: Numeric) Belian seedling plots (described in worksheet Belian seedlings)Description: Belian seedlings were surveyed in 540 5x5m subplots in the centre zone of the 4-ha plot. Each row corresponds to a seedling plot. In each seedling plot, healthy and damaged (primarily herbivory damage by mammals, defined as damaged when top shoot, i.e. apical meristem, was missing). Other variables recorded in seedling plots include slope angle and ground vegetation cover.Number of fields: 10Number of data rows: 540Fields: X10: X coordinate of the southwest corner of 10x10 m subplot (Field type: Numeric) Y10: Y coordinate of the southwest corner of 10x10 m subplot (Field type: Numeric) healthy: count of healthy Belian seedlings within 5x5m plot. "healthy" is defined as opposed to "damaged" (see next data column) (Field type: Abundance) damaged: count of damaged Belian seedlings within 5x5m plot. "damaged" is defined as when a seedling suffered severe herbivory damage (including dead seedlings) with the apical meristem entirely missing (eaten by mammals) (Field type: Abundance) slope: mean slope angle for each seedling subplot, a measure of habitat topography (Field type: Numeric) veg_cover: ground vegetation cover for each subplot, a measure of habitat and interspecific competition (Field type: Ordered Categorical) cache: when seedlings occur in dense clusters especially under fallen tree or in buttress crevices, this may indicate they were deposited there by rodents, i.e. caching (Field type: Comments) remark: remark on habitat. seedling plots on footpath or water should be excluded from data analysis (Field type: Comments) collected_by: members of field teams: 4 groups of 4 students each (Field type: Comments) total: total number of Belian seedlings (Field type: Abundance) Date range: 2018-01-29 to 2018-02-01 Latitudinal extent: 4.7383 to 4.7383 Longitudinal extent: 116.9713 to 116.9713 Taxonomic coverage:
All taxon names are validated against the GBIF backbone taxonomy. If a dataset uses a synonym, the accepted usage is shown followed by the dataset usage in brackets. Taxa that cannot be validated, including new species and other unknown taxa, morphospecies, functional groups and taxonomic levels not used in the GBIF backbone are shown in square brackets.Plantae
- Tracheophyta
- - Magnoliopsida
- - - Laurales
- - - - Lauraceae
- - - - - Eusideroxylon
- - - - - - Eusideroxylon zwageri
 
Type Of Material Database/Collection of data 
Year Produced 2019 
Provided To Others? Yes  
 
Description Assessing Indonesia's Logged Forests in a Changing Climate 
Organisation University of Leeds
Department School of Geography Leeds
Country United Kingdom 
Sector Academic/University 
PI Contribution My postdoc is conducting the fieldwork and data collection on the project. She is leading the design and implementation of the project, with support and input from myself.
Collaborator Contribution They raised the money for the project, which has been transferred to Imperial College London to extend the contract time available to a fixed-term postdoc.
Impact Fieldwork has just started, but no data have as yet been collated or analysed
Start Year 2017