ACTIVITIES DURING SABBATICAL AT NCEAS
August 23 - December 22, 1998
I started with an extensive search of various ways in which diversity has been partitioned into spatial, structural or other components. This turned out to be laborious because of conceptual, nomenclatural and procedural ambiguities in the literature. We have reached a point in which main alternative approaches can be defined and contrasted. These are based on either:
Different goals also affect choice of methods. Most partitioning has concentrated either on teasing out richness/evenness components (Hill, Buzas and a vast literature) or on factoring out local and turnover components. A more complete approach which allows for both spatial and resource-based components has been suggested before but never fully developed. I intend to complete this revision with simulations on artificial data sets, started with Tim Blackburn, in the next few months (ref#6).
Local and regional diversity and community structure in insect-plant assemblages
The more specific part of my initial proposal was to work on the extensive data set my group and I have produced over the last four years in Brazil, on flowerhead-feeding insects associated with Brazilian Asteraceae. While I did make some progress, analysis was hampered by the practical difficulties of cross-checking dubious entries at a great distance from field books, herbaria and collections. I therefore decided to concentrate on activities which could be carried out more effectively here, as well as profit from the special facilities and personal contact at NCEAS.
An analysis of a subset of the Brazilian insect/Asteraceae data produced several new and unexpected results. For instance, we showed that generalists occur proportionally in more localities on plants with larger species assemblages, whereas specialists do the reverse. An initial report of these results is in press (Prado & Lewinsohn, ref#2).
A further approach involves a cross-continental comparison of Asteraceae and flowerhead-feeding insects in Brazil (our data) and Europe (an extensive data set produced by Helmut Zwölfer at the former IIBC). An initial comparison was done by path analysis, but there are some limitations which seemingly can be overcome in structural equation modeling. This is also to be completed in 1999 (ref #7).
The necessity for inventorying and investigating many groups of organisms through resource-centered inventories, and some appropriate methods, are discussed in Lewinsohn et al. (ref. #5); a draft is being circulated for comments. This applies to endophagous and ectophagous herbivorous insects, mites and nematodes as well as to animal parasites; also to mycorrhyzal fungi among other groups.
Interactions between host ranges and species assemblages
These are being summarized in a short paper with John Lawton, which stresses the inherent limitations of analyzing either insect host ranges (the "entomocentric"approach), or host-associated assemblages (a "phytocentric" approach) on their own, whereas the dynamics of each of them are required to explain patterns observed in the other. An example of this is the interaction between host range and assemblage size referred to above (Prado & Lewinsohn, in press). A simple graphical model should help to clarify this and perhaps stimulate further research. A first draft is being circulated for comments and should be submitted in early 1999 (ref#4).
Seminars and interaction within NCEAS and UCSB
I gave a departmental seminar at EEMB/USCB on "Interaction diversity of insects
and host plants - Brazilian Compositae and their flowerhead feeders" on 27 October 1998.
An Eco-lunch talk on "Spatial and interaction components of diversity measures - difficulties and possible solutions" was presented at NCEAS on 12 November 98.
Interactions at NCEAS were extensive, diversified and very rewarding; this included other sabbatical fellows, postdocs, participants at workshops. During these 4 months I attended/took part in some sessions of five workshops and profited very much from contact with new ideas and ongoing work.
Interaction with faculty in the UCSB campus was less extensive and restricted to specific contacts based on common interest; however, I did attend many departmental seminars as well as course seminars.
Publications, editorial work
Lewinsohn and Martins, 1999 (ref#1). This edited volume on Brazilian research on insect ecology, evolution and behavior was to have been completed in early 98, but many chapters were late (including my own) so I did a substantial amount of editing here in September. NCEAS support is acknowledged in the Preface. I also co-authored two chapters, below.
Lewinsohn and Vasconcellos (ref #2) is a summary of behavioral adaptations of insects that feed on laticiferous plants, and was largely written at NCEAS.
Prado and Lewinsohn (ref #3) was completed and revised here. It presents an analysis of the interactions between resource specialization and geographical distribution on a subset of our data on insects on Brazilian Asteraceae; it represents an advance instalment of the more extensive analyses on which I started work at NCEAS (ref#8).
International Congress of Entomology 2000
As member of the Organizing Committee, I co-organized the symposia of the section of Biodiversity and Biogeography. I hope to include Camille Parmesan's work on species range shifts, done at NCEAS, in one of the key symposia.
Diagnosis of the state of knowledge of biodiversity in Brazil
This is part of a commitment with the Brazilian Ministry of the Environment, produced under contract with the UN. Ongoing coordination of this project was continued while at NCEAS and will be acknowledged.
Publications derived from activities at NCEAS
Manuscripts to be submitted in early 1999
Work and manuscripts to complete in 1999
Prospects for future activities with NCEAS
Improving estimates of global diversity
A workgroup proposal for mid-1999 to 2001 is being planned.
A proposal for a workgroup is to be submitted jointly with Jordi Bascompt (attached draft)
I plan on offering a contribution to the workgroup led by Mike Willig on latitudinal gradients, based on the temperate/tropical comparison that I am doing on flowerhead-feeders; some other groups of arthropods offer promising possibilities of analysis.